This protocol will examine the hypothesis that level of antibody to gB and neutralizing antibody level are associated with decreased rate of congenital CMV infection and decreased virulence of fetal infection. This is part of a program project grant, "Perinatal infections, immunity and maldevelopment", HD10699019. The GCRC has provided personnel, facilities and support for clinical assessments since the initial funding of this program project and continues to provide support that is absolutely essential for collection of the clinical data. The specific aims which follow will require correlation of data on the outcome of congenital CMV infection with laboratory studies of maternal immune response to specific CMV proteins. (1) Define the natural history of central nervous system sequelae in children with congenital CMV infection through serial medical, audiological, visual and neuropsychological evaluations. (2) Among mothers who are immune to CMV prior to conception, determine whether preconceptional maternal serum level of virus neutralizing antibody and level of antibody to CMV envelope glycoprotein B are associated with transmission of virus to the fetus. (3) Compare maternal preconceptional antibody levels (gB and neut antibody) in patients with congenital CMV infection born to the mother's studies in aim 2. (4) Among mothers with primary CMV infection during pregnancy, determine whether level of neutralizing antibody and level of antibody to envelope glycoprotein B are associated with risk of transmission of virus to the fetus. (5) Among congenitally infected offspring of mothers studies in aim 4, compare maternal prenatal level of antibody (gB and neut) of patients with CNS sequelae with that of patients who remain sequelae free.